Video identification blade form and clearance method based on multiple identification points of blade

By arranging multiple marking points on the blades of the wind turbine set and fitting the blade morphological curve using video recognition technology, the problem of difficulty in identifying blade morphological changes and load conditions in the prior art is solved, and more accurate clearance monitoring and early warning functions are achieved , ensure the safe and stable operation of the wind turbine.

CN119062523AActive Publication Date: 2024-12-03CGN GREAT WUYANGPING WIND POWER CO LTD

Patent Information

Application Number
CN202411088263.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-12-03
Estimated Expiration
2044-08-09

AI Technical Summary

Technical Problem

The existing wind turbine blade morphology monitoring system is difficult to accurately identify the blade morphology changes and load conditions, especially when the blade is deformed, it cannot effectively identify the blade tip part, resulting in inaccurate clearance monitoring.

Method used

Multiple marking points are arranged on the blade, and the blade morphology and clearance method are identified through video. The blade morphology curve is fitted using the distance information between the marking points and the leaf root and the tower, and the load condition and clearance value are calculated.

Benefits of technology

Through the video recognition method of multiple marking points, the blade morphology changes and load conditions can be obtained more accurately, the reliability of the net space value is improved, and it can be used for the start of early warning and clearance protection strategies to ensure the safe operation of the wind turbine.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119062523A_ABST
    Figure CN119062523A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of wind generating set blade form and clearance monitoring, in particular to a video blade form and clearance identification method based on multiple identification points of a blade. A camera is arranged at the position, close to a blade, of a cabin, n identification points are arranged on the blade, a video that the blade passes through a tower is shot, the plane position of each identification point is recognized and calculated according to video information, a blade form curve is fitted, the load condition is calculated, the blade tip position is calculated according to the blade form curve, and a clearance value is calculated. According to the method, a plurality of identification points are arranged on a blade, a blade form curve is fitted according to the identification points and position information of a blade root, and the form change condition and the load condition of the blade are obtained; meanwhile, the clearance value is obtained through the blade form curve, and the clearance value is more accurate and reliable.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of blade morphology and clearance monitoring of wind turbine generator sets, and in particular to a method for identifying blade morphology and clearance based on a video of multiple identification points of a blade. Background Art

[0002] In pursuit of low electricity costs, large-scale wind turbines have become the mainstream trend in the current wind power industry. The increase in wind turbine capacity, larger models, and longer blades have also brought more challenges to the safe and stable operation of the units, such as the risk of wind turbine tower sweeping. In order to ensure the safe operation of the unit, the clearance monitoring system and blade deformation monitoring system for wind turbines are increasing. At present, the more common clearance monitoring systems include clearance monitoring systems based on image technology methods to identify blade tips, laser clearance monitoring systems based on distance sensors, millimeter wave radar clearance monitoring systems, etc.; blade deformation monitoring includes blade internal images based on image technology methods, blade root blade images, etc., as well as blade tip displacement identification based on distance sensors, blade internal wires, blade root strain gauges, etc.

[0003] For example, Chinese patent CN116816613A discloses a blade morphology monitoring system for a wind turbine generator set. By installing a first image capture device, a second image capture device, and a third image capture device on the outside of the hub cover, the image capture device rotates with the wind wheel and remains relatively still with the blade. The blade morphology, hub speed, and blade pitch angle parameters during the operation of the wind turbine generator set can be obtained by image recognition, and then the wind wheel load, blade and tower clearance can be monitored to ensure the safe operation of the unit. However, the image capture device is installed on the outside of the hub cover, and it is necessary to obtain the hub cover speed and blade pitch angle information in real time. Moreover, the blade tip part may be blocked in the picture due to blade deformation at this viewing angle. Under the condition of partial load of the blade, the blade morphology cannot be identified and the morphological changes of the blade cannot be understood. Chinese patent CN117028161A discloses a blade clearance detection system, method, device, and storage medium for a wind turbine generator set, including a deformation detection unit and a clearance detection unit, but its deformation detection unit can only detect the deformation data of the blade root and cannot identify the entire blade morphology. Summary of the invention

[0004] The present invention provides a method for identifying blade morphology and clearance based on video of multiple identification points of the blade. The method arranges multiple identification points on the blade, fits the blade morphology curve according to the position information of each identification point and the blade root, and obtains the morphological change and load conditions of the blade; at the same time, the clearance value is obtained through the blade morphology curve, and the clearance value is more accurate and reliable.

[0005] In order to solve the above technical problems, the technical solution of the present invention is:

[0006] A method for identifying blade shape and clearance based on a video of multiple identification points of a blade comprises the following steps:

[0007] S1. Deploy a camera near the blade in the cabin, and arrange marking points at n+1 equally divided points between the blade root and the blade tip. Arrange n marking points on each blade, and all marking points are located on the leeward side of the blade. Record the distance r between each marking point and the blade root in turn.

[0008] S2, shooting a video of the blade passing through the tower, identifying and calculating the minimum distance d between each identification point and the tower according to the video information;

[0009] S3, calculating the plane position of each identification point according to the distance r between each identification point and the blade root and the minimum distance d between the corresponding identification point and the tower;

[0010] S4, fitting the blade shape curve and calculating the load condition according to the plane position of each identification point and the blade model;

[0011] S5. Calculate the tip position of the leaf according to the leaf shape curve;

[0012] S6. Calculate the clearance value based on the blade tip position and fan structural parameters.

[0013] There are no less than five marking points on each leaf.

[0014] In S1, the distance between each identification point and the blade root is calculated according to the blade length.

[0015] The specific calculation method of the minimum distance d between each identification point and the tower in S2 is:

[0016] S2.1. After the camera is installed, calibrate the conversion ratio between the pixel distance in the video image and the actual object distance;

[0017] S2.2, according to the video information, identify the pixel position of the identification point on the blade in each frame image when the blade passes through the tower, obtain the pixel position of each identification point closest to the tower by fitting the curve, and identify the pixel position of the tower in the horizontal direction corresponding to each identification point when it is closest to the tower, so as to obtain the minimum pixel distance between each identification point and the tower;

[0018] S2.3. According to the conversion ratio between pixel distance and actual distance, the minimum pixel distance between each identification point and the tower is converted into the minimum distance d between each identification point and the tower.

[0019] The specific calculation method of the plane position of each identification point in S3 is:

[0020] S3.1. With the blade root position as the origin 0, establish the x-axis coordinate in the horizontal direction from the head of the nacelle to the tail, and establish the y-axis coordinate in the vertical direction;

[0021] S3.2. According to the structural parameters of the wind turbine, the horizontal coordinate of the edge point of the tower close to the blade is obtained as t0;

[0022] S3.3, Assume that the coordinates of the i-th identification point are (x i ,y i ), i is an integer from 1 to n, and the distance between the i-th identification point and the leaf root is r i , then x i 2 +y i 2 =r i 2 ;

[0023] S3.4. It is known that the minimum distance between the i-th identification point and the tower is d i , then x i =t0-d i ;

[0024] S3.5. The coordinates of the i-th identification point are calculated as Thus, the coordinates of each marking point are obtained.

[0025] The methods used to fit the leaf morphology curve in S4 are linear regression, polynomial regression, local weighted scatter point smoothing, logistic regression, and nonlinear regression methods.

[0026] According to the blade morphology and vibration mode, the polynomial regression method is used to fit the blade morphology curve. The specific steps are as follows:

[0027] S4.1, using the third-order polynomial regression method, then y = w0 + w1x + w2x 2 +w3x 3 ;

[0028] S4.2. Given the blade root position and the plane positions of n identification points, the optimal parameters w0, w1, w2, w3 are calculated using the scattered point minimization error method to obtain the blade morphology curve.

[0029] The specific calculation method of the load condition in S4 is: after obtaining the blade shape curve, the blade shape curve is compared with the shape of the blade in the non-stress state to obtain the deformation of each point, and then the current shape load condition of the blade is obtained.

[0030] In S5, the position of the leaf tip is calculated according to the distance between the leaf tip and the leaf root and the fitted leaf morphology curve.

[0031] The clearance value and blade shape changes are fed back to the main control, which determines whether an early warning is needed and initiates a clearance protection strategy.

[0032] Compared with the prior art, the present invention has the following beneficial effects: (1) In the present invention, multiple identification points are arranged on the blade, and the minimum distance between each identification point and the tower when the blade passes through the tower is identified by means of video images, so as to obtain the planar position of each identification point when the blade is closest to the tower, and then the blade shape curve when the blade is closest to the tower is obtained by fitting according to the planar position of each identification point, the position information of the blade root and the structural parameters of the wind turbine. By comparing the blade shape curve with the shape of the blade in a non-stressed state, the shape change of the blade is obtained, and then the current shape load of the blade is obtained; at the same time, the blade tip position and the clearance value are calculated through the blade shape curve.

[0033] (2) In the present invention, the clearance value is calculated by identifying the positions of multiple marking points and fitting the blade shape, so the clearance value is more accurate and reliable.

[0034] (3) In the present invention, the main control can issue an early warning and start the clearance protection strategy according to the clearance value, load condition and deformation condition to protect the safety of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of a method for recognizing blade shape and clearance based on a video of multiple identification points of a blade according to the present invention;

[0036] Figure 2 It is a schematic diagram of the synthesis of the video image trajectory when the blade passes through the tower in the present invention;

[0037] Figure 3 A schematic diagram of establishing a coordinate system with the blade root as the origin in the present invention;

[0038] In the figure: 1-blade root, 2-blade tip, 3-camera, 4-marking point, 5-fitting blade shape curve, 6-tower. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0040] The present invention provides a method for identifying blade shape and clearance based on a video of multiple blade identification points 4, such as Figure 1 As shown, the following steps are included:

[0041] S1. Deploy a camera 3 near the blade in the cabin, and arrange marking points 4 at n+1 equally divided points between the blade root 1 and the blade tip 2. Arrange n marking points 4 on each blade, and all the marking points are located on the leeward side of the blade, that is, the side facing the tower in the normal power generation state (about 0 degree pitch angle). Record the distance r between each marking point 4 and the blade root 1 in turn, that is, the distance between each marking point 4 and the blade root 1 is r1, r2...r n ;

[0042] Specifically, the distance between each identification point 4 and the blade root 1 can be calculated according to the blade length. It is known that n identification points 4 are arranged at n+1 equally divided points between the blade root 1 and the blade tip 2. If the length of the blade (i.e., the distance from the blade root 1 to the blade tip 2) is L, then the distance of each equally divided point is L / (n+1), and the distance r between each identification point 4 and the blade root 1 is L / (n+1), 2L / (n+1)...nL / (n+1) in sequence;

[0043] Specifically, there are no less than five marking points 4 on each blade to ensure the accuracy of the later blade shape fitting. In this embodiment, there are five marking points 4 on each blade. Figure 1 , the distances between the five marking points 4 and the blade root 1 are r1=L / 6, r2=L / 3, r3=L / 2, r4=2L / 3, r5=5L / 6 respectively; further, the five marking points 4 are marking stickers of different colors, such as purple, blue, green, orange, and red marking stickers, which are convenient for image recognition;

[0044] S2, shoot a video of the blade passing through the tower 6, identify and calculate the minimum distance d between each mark point 4 and the tower 6 according to the video information, that is, the minimum distance d between each mark point 4 and the tower 6 is d1, d2...d n ;

[0045] The specific calculation method of the minimum distance d between each identification point 4 and the tower 6 is:

[0046] S2.1. After the camera 3 is installed, the conversion ratio between the pixel distance in the video image and the physical distance is calibrated; specifically, during the calibration, the pixel position and the physical position of the identification point 4 are used as calibration points. Since the physical distance between two adjacent identification points 4 is known, and the pixel distance between two adjacent identification points 4 can be obtained through video information, the conversion ratio between the pixel distance in the video image and the physical distance is obtained by calibration;

[0047] S2.2, according to the video information, the pixel position of the identification point 4 on the blade in each frame image when the blade passes through the tower 6, the pixel position of each identification point 4 closest to the tower 6 is obtained by fitting the curve, and the pixel position of the tower 6 in the horizontal direction corresponding to each identification point 4 when the identification point 4 is closest to the tower 6 is identified, so as to obtain the minimum pixel distance between each identification point 4 and the tower 6;

[0048] In this embodiment, based on historical experience, the curve of the blade identification point 4 passing through the tower 6 in each frame image is fitted into a straight line, such as Figure 2 As shown, the identification point 4 is fitted by a straight line fitting method, so as to obtain the pixel position of each identification point 4 closest to the tower 6. However, according to the on-site working conditions of the wind turbine, other fitting methods are not excluded;

[0049] S2.3, according to the conversion ratio between pixel distance and physical distance, convert the minimum pixel distance between each identification point 4 and the tower 6 into the minimum distance d between each identification point 4 and the tower 6;

[0050] In this embodiment, the minimum distances between the five identification points 4 and the tower 6 are d1, d2, d3, d4, and d5 respectively;

[0051] S3, calculating the plane position of each identification point 4 according to the distance r between each identification point 4 and the blade root 1 and the minimum distance d between the corresponding identification point 4 and the tower 6;

[0052] The specific calculation method of the plane position of each identification point 4 is:

[0053] S3.1. With the blade root 1 as the origin 0, establish the x-axis coordinate in the horizontal direction from the head of the nacelle to the tail, and establish the y-axis coordinate in the vertical direction, such as Figure 3 As shown;

[0054] S3.2. According to the structural parameters of the wind turbine, the horizontal coordinate of the edge point of the tower 6 close to the blade is t0. Figure 3 ;

[0055] S3.3, Assume that the coordinates of the i-th identification point 4 are (x i ,y i ), i is an integer from 1 to n, and the distance between the i-th identification point 4 and the blade root 1 is known to be r i , ignoring the large deformation of the marker point 4, then x i 2 +y i 2 =r i 2 ;

[0056] S3.4. It is known that the minimum distance between the i-th identification point 4 and the tower 6 is di At this time, the coordinates of the edge point of the tower 6 close to the blade corresponding to the i-th identification point 4 in the horizontal direction are (t0, y i ), then x i =t0-d i ;

[0057] In this embodiment, the coordinates of the edge points of the tower 6 close to the blades in the horizontal direction corresponding to the five identification points 4 are (t0, y1), (t0, y2), (t0, y3), (t0, y4), and (t0, y5), respectively. Figure 3 ;

[0058] S3.5. The coordinates of the i-th identification point 4 are calculated as Thus, the coordinates of each identification point 4 are obtained;

[0059] S4, fitting the blade shape curve 5 and calculating the load condition according to the plane position of each identification point 4 and the blade model;

[0060] Specifically, the method of fitting the blade morphology curve 5 includes linear regression, polynomial regression, local weighted scatter point smoothing, logistic regression, nonlinear regression and the like. In this embodiment, the polynomial regression method is used to fit the blade morphology curve 5 according to the blade morphology and vibration mode, and the specific steps are as follows:

[0061] S4.1, taking the third-order polynomial regression method as an example, then y=w0+w1x+w2x 2 +w3x 3 ;

[0062] S4.2, given the position of the blade root 1 and the plane positions of the n identification points 4, substitute the coordinates of the n+1 points into the above formula, and according to the method of minimizing the error of scattered points, find the optimal parameters w0, w1, w2, w3 to obtain the blade morphology curve;

[0063] After obtaining the blade shape curve, compare the blade shape curve with the shape of the blade in the non-stress state to obtain the deformation of each point, and then obtain the current shape load of the blade;

[0064] S5, calculating the position of the blade tip 2 according to the blade shape curve;

[0065] Specifically, the position of the blade tip 2 is calculated according to the distance L between the blade tip 2 and the blade root 1 and the fitted blade shape curve 5;

[0066] Assume that the coordinates of blade tip 2 are (x tip ,y tip ),but:

[0067] x tip 2 +ytip 2 =L 2 ;

[0068] y tip =w0+w1x tip +w2x tip 2 +w3x tip 3 ;

[0069] From the previous steps, L, w0, w1, w2, and w3 are known, and the coordinates of the blade tip 2 can be calculated by solving the equation;

[0070] S6. Calculate the clearance value D according to the position of the blade tip 2 and the structural parameters of the wind turbine. According to the structural parameters of the wind turbine, the horizontal coordinate of the edge point of the tower 6 close to the blade is known to be t0. According to the horizontal coordinate x of the blade tip 2 obtained in S5, tip , the calculated clearance value D is x tip -t0.

[0071] In the present invention, multiple identification points 4 are arranged on the blade, and the minimum distance between each identification point 4 and the tower 5 when the blade passes through the tower 6 is identified by video images, so as to obtain the plane position of each identification point 4 when the blade is closest to the tower 6, and then the blade shape curve when the blade is closest to the tower 6 is obtained by fitting according to the plane position of each identification point 4, the position information of the blade root 1 and the structural parameters of the wind turbine. The deformation of the blade is obtained by comparing the blade shape curve with the shape of the blade in the non-stressed state, and then the current shape load of the blade is obtained; at the same time, the blade tip position and the clearance value are calculated through the blade shape curve. The clearance value, blade shape change and load condition are fed back to the main control, and the main control determines whether an early warning is needed and starts the clearance protection strategy.

Claims

1. A method for identifying blade shape and clearance based on a video of multiple identification points of a blade, characterized in that The following steps are involved: S1. Deploy a camera near the blade in the cabin, and arrange marking points at n+1 equally divided points between the blade root and the blade tip. Arrange n marking points on each blade, and all marking points are located on the leeward side of the blade. Record the distance r between each marking point and the blade root in turn. S2, shooting a video of the blade passing through the tower, identifying and calculating the minimum distance d between each identification point and the tower according to the video information; S3, calculating the plane position of each identification point according to the distance r between each identification point and the blade root and the minimum distance d between the corresponding identification point and the tower; S4, fitting the blade shape curve and calculating the load condition according to the plane position of each identification point and the blade model; S5. Calculate the tip position of the leaf according to the leaf shape curve; S6. Calculate the clearance value based on the blade tip position and fan structural parameters.

2. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: There are no less than five marking points on each leaf.

3. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: In S1, the distance between each identification point and the blade root is calculated according to the blade length.

4. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: The specific calculation method of the minimum distance d between each identification point and the tower in S2 is: S2.

1. After the camera is installed, calibrate the conversion ratio between the pixel distance in the video image and the actual object distance; S2.2, according to the video information, identify the pixel position of the identification point on the blade in each frame image when the blade passes through the tower, obtain the pixel position of each identification point closest to the tower by fitting the curve, and identify the pixel position of the tower in the horizontal direction corresponding to each identification point when it is closest to the tower, so as to obtain the minimum pixel distance between each identification point and the tower; S2.

3. According to the conversion ratio between pixel distance and actual distance, the minimum pixel distance between each identification point and the tower is converted into the minimum distance d between each identification point and the tower.

5. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: The specific calculation method of the plane position of each identification point in S3 is: S3.

1. With the blade root position as the origin 0, establish the x-axis coordinate in the horizontal direction from the head of the nacelle to the tail, and establish the y-axis coordinate in the vertical direction; S3.

2. According to the structural parameters of the wind turbine, the horizontal coordinate of the edge point of the tower close to the blade is obtained as t0; S3.3, Assume that the coordinates of the i-th identification point are (x i ,y i ), i is an integer from 1 to n, and the distance between the i-th identification point and the leaf root is r i , then x i 2 +y i 2 =r i 2 ; S3.

4. It is known that the minimum distance between the i-th identification point and the tower is d i , then x i =t0-d i ; S3.

5. The coordinates of the i-th identification point are calculated as Thus, the coordinates of each marking point are obtained.

6. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: The methods used to fit the leaf morphology curve in S4 are linear regression, polynomial regression, local weighted scatter point smoothing, logistic regression, and nonlinear regression methods.

7. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1 is characterized in that: According to the blade morphology and vibration mode, the polynomial regression method is used to fit the blade morphology curve. The specific steps are as follows: S4.1, using the third-order polynomial regression method, then y = w0 + w1x + w2x2 + w3x3; S4.

2. Given the blade root position and the plane positions of n identification points, the optimal parameters w0, w1, w2, w3 are calculated using the scattered point minimization error method to obtain the blade morphology curve.

8. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1, characterized in that: The specific calculation method of the load condition in S4 is: after obtaining the blade shape curve, the blade shape curve is compared with the shape of the blade in the non-stress state to obtain the deformation of each point, and then the current shape load condition of the blade is obtained.

9. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1, characterized in that: In S5, the position of the leaf tip is calculated according to the distance between the leaf tip and the leaf root and the fitted leaf morphology curve.

10. The method for recognizing blade shape and clearance based on a video of multiple blade identification points according to claim 1, characterized in that: The clearance value and load condition are fed back to the main control, which determines whether an early warning is needed and initiates the clearance protection strategy.

Citation Information

Patent Citations

  • Method and device for determining tower clearance of wind generating set

    CN111255636A

  • Unmanned aerial vehicle-based fan blade and tower drum inspection identification system and method

    CN113759960A

  • Wind turbine generator tower clearance monitoring method and computer program product

    CN114623049A

  • Method and device for measuring clearance distance of wind driven generator based on machine vision

    CN115218801A

  • Method for measuring clearance of wind turbine generator set on ground based on machine vision

    CN115713558A

Cited By

  • Fan blade clearance monitoring method and system based on OFDR

    CN120798686A